Project Summary: ATF3 is a highly conserved transcription factor that is ubiquitously upregulated in the mammalian response to DNA damage. Preliminary data shows that ATF3 is significantly upregulated at both the mRNA and protein levels in many human breast tumors. Overexpression of ATF3 in keratin 5-expressing cells of BK5.ATF3 transgenic mice, including myoepithelial cells of the mammary gland, results in a high incidence of mammary carcinomas that overexpress p-catenin, K5, K6 and K8, similar to tumors induced by genetic alterations of the Wnt/p-catenin pathway. It is hypothesized that the oncogenic effects of ATF3 expression in this model involve upregulation of the Wnt/p-catenin signaling pathway, and experiments are proposed to test this hypothesis. In Specific Aim 1, the TOPGAL reporter gene will be used to determine whether Wnt/p-catenin signaling is active in mammary tumors arising in BK5.ATF3 transgenic mice. These tumors will also be characterized at the mRNA level for activation of the Wnt/p-catenin and other important signaling pathways. In Specific Aim 2, premalignant, metaplastic lesions arising in virgin mammary glands of BK5-ATF3 transgenic mice will also be examined molecularly for evidence that the Wnt/p-catenin pathway is upregulated, using the same techniques. Laser capture microdissection of frozen sections will be used to isolate the lesions for RNA preparation. In Specific Aim 3, transgenic mice will be developed in which Wnt/p- catenin signaling in K5-expressing cells can be conditionally eliminated. This genetic alteration will be combined with the BK5.ATF3 transgene and the effect on development of both squamous metaplastic lesions and adenosquamous carcinomas will be determined. In Specific Aim 4, the proportion of putative stem/progenitor cell populations in the glands of wild-type and transgenic mice will be assayed. The effect of loss of the Sdd gene, known to be required for Wnt/p-catenin-induced mammary tumors, on ATF3-induced tumorigenesis will be determined. The results of these studies are expected to provide clues to the mechanism that links ATF3 with Wnt/p-catenin signaling and to suggest new mechanisms for the involvement of environmental carcinogens with human tumorigenesis. This will establish relevance of ATF3 to several tumor types with known relationships to Wnt signaling, and particularly to basal-like human breast tumors, a subset with poor prognosis and phenotypic similarity to ATF3-induced mouse mammary tumors. Relevance: Breast cancer afflicts 1 in 9 American women. Only about 10% of these cancers are explained by inherited cancer genes; the rest are likely due to exposure to DMA-damaging environmental factors and genetic susceptibility factors. The known importance of ATF3 in the response to DNA damage and the oncogenic properties of ATF3 in the current model may provide a mechanistic link between environmental exposures and human breast cancer susceptibility, especially for basal-like tumors with poor prognosis.